Open dlschwartz opened 1 year ago
FYI, @dlschwartz the index created from the CSV transform (which we might have added to GitHub manually?) is created using these two functions: https://github.com/wlpotter/csv-to-srophe/blob/main/modules/csv2subjects.xqm#L129-L174.
If I remember right, we had a 'blank' template of the index which has the different listUri elements, which it uses to collect the skos:broader
matches.
It looks like we treated 'relationships' separately (https://github.com/wlpotter/csv-to-srophe/blob/main/modules/csv2subjects.xqm#L146-L161) -- I remember this being tricky to get right.
Let me know if you'd like me to dig in further to see what the difference between what this script and the one in the App run on. If memory serves, the XQuery here wasn't used to create the Taxonomy index, but there was instead an old XSLT that did it?
I'd also be happy to meet and talk through the CSV transform's functions here, if that would be helpful
I need to check the data in the recently-generated taxonomy xml and compare it to the data in the outdated taxonomy. This index needs to reflect the hierarchies in the xml so that SPEAR can accurately validate against the data there.
Also, let's rename this taxonomyIndexSPEAR.